See data and maps.

Plain text

Ochs, S. (2021). “Prohibitive and its relation with the imperative”. In: Typological Atlas of the Languages of Daghestan (TALD). Ed. by M. Daniel, K. Filatov, T. Maisak, G. Moroz, T. Mukhin, C. Naccarato and S. Verhees. Moscow: Linguistic Convergence Laboratory, NRU HSE. DOI: 10.5281/zenodo.6807070. http://lingconlab.ru/dagatlas.

BibTeX

@incollection{ochs2021,
  title = {Prohibitive and its relation with the imperative},
  author = {Samira Ochs},
  year = {2021},
  editor = {Michael Daniel and Konstantin Filatov and Timur Maisak and George Moroz and Timofey Mukhin and Chiara Naccarato and Samira Verhees},
  publisher = {Linguistic Convergence Laboratory, NRU HSE},
  address = {Moscow},
  booktitle = {Typological Atlas of the Languages of Daghestan (TALD)},
  url = {http://lingconlab.ru/dagatlas},
  doi = {10.5281/zenodo.6807070},
}

1 Introduction

The prohibitive is typically defined as the negative imperative, i.e. it “implies making someone not do something, having the effect of forbidding, preventing, or restricting” (Aikhenvald, 2017: 3). This chapter focuses on the formation of the prohibitive in the languages of Daghestan and neighboring regions, analyzing two different aspects of the morphological coding: first, the verb form (especially whether it is an imperative form or not), and second, the type of negation marker/affix used. Based on this, the general encoding types are deduced. Additionally, the phonological form of the markers is shortly analyzed.

In the languages of the world, the most common strategy to code the prohibitive is to use the second (singular or plural) imperative and a sentential negative strategy not found in indicative declaratives (37% of the sample in the chapter on prohibitives in WALS, see van der Auwera & Lejeune (2013)). Two other common strategies involve the use of non-imperative verbal constructions and a negation marker not found in declaratives (29%), as well as the use of the imperative combined with a negation found in declaratives (23%). The use of a non-imperative verbal construction together with a negative strategy found in declaratives is the least common strategy in the languages of the world (11%).

Interestingly, “[c]ategories expressed in positive imperatives tend to be neutralized under negation” (Aikhenvald, 2017: 19). For instance, some languages of the TALD sample differentiate between transitivity and intransitivity in positive imperatives but allow only one form for the prohibitive. Likewise, only one stem type is used for prohibitives if the language has both a perfective and an imperfective stem.

While the variation of stem types is only superficially coded in the WALS (i.e. normal imperative vs. special imperative; see van der Auwera & Lejeune (2013), the types of prohibitive markers are not considered at all. For the TALD sample, the stem types are examined regarding their ‘imperativity’. Furthermore, the different types of affixes and particles used as prohibitive markers are indexed. Our goal is to classify the languages according to the verb form and negation marker used to encode the prohibitive.

2 Results

2.1 Non-imperative verb form + special prohibitive marker

Most of the sampled languages make use of the second most common strategy accounted for in the WALS: a special, non-imperative verb form combined with a special prohibitive marker. 28 out of 37 languages belong to this category (75.68 %). Most languages make use of suffixes (16), followed by circumfixes (6), prefixes (5), and infixes (1).

The types of verb forms used for the prohibitive are much less clear in the literature. We find detailed descriptions of imperfectivity vs. perfectivity (e.g. (Kibrik, 1977), (Kibrik, 1999)), unspecified reference to “stem” or “root” (e.g. (Maxmudova, 2002), (Žirkov, 1955)), and sometimes no information at all (e.g. (Magomedova, 2004)). A detailed investigation of these different underlying verb forms is a desideratum for further in-depth analysis.

The following 28 languages belong to the category ‘non-imperative verb form + special prohibitive marker’:

  • All 9 Avar-Andic languages: Northern Akhvakh, Andi, Standard Avar, Bagvalal, Botlikh, Chamalal, Godoberi, Karata, Tindi
  • 5 Dargwa varieties: Standard Dargwa, Mehweb, Itsari, Kubachi, Tanty
  • Lak
  • 8 Lezgic languages (except for Udi): Agul, Archi, Budukh, Kryz, Lezgian, Rutul, Tabasaran, Tsakhur
  • All 5 Tsezic languages: Bezhta, Hinuq, Hunzib, Khwarshi, Tsez

While the suffixes and prefixes encode only prohibition, some additional remarks are necessary about the circumfixes used in Agul, Kryz, Lak, and Tabasaran. The prefixal part of the circumfix (mV- in all four languages) is the specialized prohibitive marker, while the functions of the suffixal parts are more diverse. In Lak, the suffixes -ara and -ari mark prohibition and, additionally, second person singular (1) or plural (2).

  1. Lak (Žirkov, 1955: 119)
    ma-čič-ara
    proh-write-proh.sg
    ‘Don’t write! (Sg.)’
  2. Lak (Ibid.)
    ma-čič-ari
    proh-write-proh.pl
    ‘Don’t write! (Pl.)’

In Agul (3)-(4) and Tabasaran (3), the suffixes -an and -an-aj are less transparent. The imperative used to be formed by attaching these suffixes to the verb stem; in contemporary Agul and Tabasaran, however, the second person singular imperative is reduced to the bare stem, and the second person plural is marked by -aj. The marker -an is lost in both forms. In the prohibitive, -an is still obligatory, both in the singular and in the plural. This leads to two possible interpretations: either the prohibitive is formed by prefixing the traditional imperatives with mV-, or the prohibitive has a specialized circumfixal marker mV-…-an/-an-aj which is attached to the bare stem. I opt for the second interpretation, as the contemporary languages use -an only for the prohibitive, i.e. the functional range of -an seems to have narrowed down to negative contexts.

  1. Agul (Magometov, 1970: 141)
    ma-χur-an
    proh-read-proh
    ‘Don’t read!’
  2. Agul (Ibid.)
    ma-χur-an-aj
    proh-read-proh-pl
    ‘Don’t read!’
  3. Tabasaran (Mixail E. Alekseev & Šixalieva, 2003: 70)
    mu-ulup-an
    proh-show-proh
    ‘Don’t show!’
  4. Tabasaran (Ibid.)
    mu-ulup-an-aj
    proh-show-proh-pl
    ‘Don’t show!’

In Kryz, both transitive and intransitive verbs are marked for gender and number. In (7), the transitive prohibitive manzu has the feminine marker -u. In (8), the transitive prohibitive muts’uvay is marked for human plural (HPL) with -vay.

  1. Kryz (Authier, 2009: 158)
    q’irš.a-k ča-da-b-ha buluša ma-nzu!
    mud-sub pv-neg-f-stain.pf(ptcp) dress proh-wash
    ‘Don’t wash the dress which is not mud stained!’
  2. Kryz (Authier, 2009: 159)
    lu riș-i gardan ǧe-t’-iz muts’u-vay!
    this girl-gen neck pv-strike-inf proh.give-hpl
    ‘Don’t let this girl’s head be cut off!’

If there is a preverb, the prefixal marker ma- is inserted between the preverb and the stem, blurring the lines between prefix and infix (e.g. ǧi-ma-șu ‘Don’t wear’; (Authier, 2009: 158). This is also true for the two Dargwa varieties Sanzhi and Tanty (Forker, 2020: 298–299), (Sumbatova & Lander, 2014: 113) as well as for the Rutul varieties Luchek and Mukhad (Mikhail E. Alekseev, 1994: 234), (Maxmudova, 2002: 170), (Ibragimov, 2004: 107). Talibov (2007: 216) classifies -mV- in Budukh as an infix. However, as he explains that it is placed between the preverb and the root, I re-interpret it as a prefix in accordance with the other languages showing such structures. The same is true for Tsakhur, where the prohibitive morpheme stands between the preverb and the class marker (Kibrik, 1999: 84, 280). A real infix can only be found for Kubachi, where the prohibitive morpheme is inserted between the reduplicated roots (Magometov, 1963: 233).

2.2 Regular imperative + special prohibitive marker

The second most common pattern in the sample languages is the combination of a regular imperative with a specialized prohibitive marker. However, this pattern is infrequent compared to the one previously described: Only 16.22 % of the languages make use of it.

The six languages in this category are Khinalug, Udi of the Lezgic branch, all three Nakh languages (Tsova-Tush, Ingush, Standard Chechen), and Armenian (an Indo-European language).

Affixes and separate particles are used as prohibitive markers: Udi uses a prefix, Khinalug has a suffix, and the other four languages use separate particles,1 e.g. Tsova-Tush (9). Khinalug is the only language in the sample that has a single affix to mark both prohibition and number. The suffixes -kuj (Sg) and -kun (Pl) are attached to the regular imperative (7) and serve as a portmanteau morpheme (Dešeriev, 1959: 116–117), (Kibrik, Kodzasov, & Olovjannikova, 1972: 191).

  1. Tsova-Tush (Dešerieva, 1999: 170)
    ma vašlib
    proh shave.imp
    ‘Don’t shave!’
  2. Khinalug (Dešeriev, 1959: 116)
    ka! ka-kuj! ka-kun!
    do do-proh.sg do-proh.pl
    ‘Do it! Don’t do it (Sg)! Don’t do it (Pl)!’

2.3 Regular imperative + regular negation marker

All three languages from the Turkic family, i.e. Kumyk, Nogai, and Azerbaijani, use the regular imperative together with a regular negation marker to encode the prohibitive. They mark the prohibitive with the general negation suffix -ma (with different variants).

In Kumyk (11) and Nogai (12), the bare stem serves as the second person singular imperative. The form is analyzed as a regular imperative and not as a special verb form, especially in the light of the imperative/prohibitive formation in Agul and Tabasaran. In those two languages, the use of the bare stem as the second singular imperative appears to be a recent development. The original imperative with -an/-an-aj still exists. In Nogai and Kumyk, there is no alternative to the use of the bare stem as the imperative, which is why an analysis as a regular imperative is reasonable.

In Kumyk, the plural marker -gɨz/giz is attached after the negation marker (see Dmitriev (1940: 119). This is also true for Azerbaijani (-jın). For Nogai, this can be confirmed in Musaev’s dictionary (2018). The two entries cited in (13) indicate that the plural marker nɨz is used after the negation marker.

  1. Kumyk (Dmitriev, 1940: 119–120)
    bar bar-ma bar-ma-gɨz
    go go-neg go-neg-pl
    ‘Go! Don’t go! Don’t go! (Pl)’
  2. Nogai (Baskakov, 1940: 87)
    al al-ma
    take take-neg
    ‘Take! Don’t take!’
  3. Nogai (Musaev, 2018)
    sora-ma-nɨz kɨjnal--nɨz
    ask-neg-pl worry-neg-pl
    ‘Don’t ask (about our life)! Don’t worry!’

2.4 Marker forms

Following Bancel et al.’s (2021) finding that the particle *ma functions as a prohibitive/negative marker in a wide variety of genetically and geographically distant languages, the markers of the TALD sample languages were analyzed in terms of their phonological form. 21 languages have an mV marker and 16 do not. The genealogical distribution is unambiguous: All five Dargwa varieties, Armenian (the only Indo-European language in the sample), Lak, almost all Lezgic languages, all Nakh languages, and all Turkic languages have an mV marker. On the other hand, all Avar-Andic languages, Khinalug, and all Tsezic languages employ other forms. Archi (Lezgic) does not have the Lezgic branch’s default mV marker. This is not surprising, as Archi is spoken in a different area and has been in contact with Avar and Lak. Interestingly, the mV markers occur in varying positions (prefix, circumfix, infix, suffix, free particle), while markers that are not mV are exclusively suffixal.

3 Distribution

In the TALD sample, only three out of the four prohibitive encoding strategies presented in the WALS are attested. Table 1 shows a comparison of the two samples.

Table 1. Comparison of the prohibitive encoding strategies in the WALS and TALD samples

Encoding type WALS TALD sample
Regular imperative + regular negation marker 113 23% 3 8.11%
Regular imperative + special prohibitive marker 182 37% 6 16.22%
Non-imperative verb form + regular negation marker 55 11% 0 0%
Non-imperative verb form + special prohibitive marker 146 29% 28 75.68%
Total 496 100% 37 100%


The languages of the TALD sample show a clear preference for the combination of non-imperative verb forms and a specialized prohibitive marker. Suffixes are the preferred affix type, but circumfixes, infixes and prefixes are also attested.

Language group is a good predictor for the encoding strategy: all Avar-Andic, Tsezic, and Dargwa varieties, almost all Lezgic languages, and Lak make use of the ‘non-imperative verb form + special prohibitive marker’ strategy; all Nakh languages, Khinalug, one Lezgic language (Udi) and Armenian (the only Indo-European language in the sample) belong to the ‘regular imperative + special prohibitive marker’ category; only the agglutinative Turkic languages encode prohibitives with a ‘regular imperative + regular negation marker’. The marker form (mV vs. other forms) is also robustly predictable by language group. However, it is noteworthy that the pattern occurs across genetically unrelated families, supporting Bancel et al.’s (2021) findings about the proto-sapiens prohibitive particle *ma.

List of glosses

f — feminine; gen — genitive; hpl — human plural; imp — imperative; inf — infinitive; neg — negation; pf — perfect; pl — plural; proh — prohibitive; ptcp — participle; pv — preverb; sg — singular; sub — subessive

References

Aikhenvald, A. Y. (2017). Imperatives and commands: A cross-linguistic view. In A. Y. Aikhenvald & R. M. W. Dixon (Eds.), Commands. A Cross-Linguistic Typology (pp. 1–45). Oxford: Oxford University Press.
Alekseev, Mikhail E. (1994). Rutul. In R. Smeets (Ed.), The indigenous languages of the Caucasus. Volume 4. Part 2 (pp. 147–212). Delmar NY: Caravan Books.
Alekseev, Mixail E., & Šixalieva, S. X. (2003). Tabasaranskij jazyk [Tabasaran]. Moskva: Akademija.
Authier, G. (2009). Grammaire kryz. Paris: Peeters.
Auwera, J. van der, Lejeune, L., & Goussev, V. (2013). The Prohibitive. In M. S. Dryer & M. Haspelmath (Eds.), The World Atlas of Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology. Retrieved from https://wals.info/chapter/71
Bancel, P. J., l’Etang, A. M. de, & Bengston, J. D. (2021). The Proto-Sapiens Prohibitive/Negative Particle *ma. Mother Tongue, 22(1), 223–241.
Baskakov, N. A. (1940). Nogajskij jazyk i ego dialekty: Grammatika, teksty i slovarʹ [Nogai and its dialects: Grammar, texts and dictionary]. Moskva: Izdatelʹstvo Akademii nauk SSSR.
Dešeriev, J. D. (1959). Grammatika xinalugskogo jazyka [Khinalug grammar]. Moskva: Akademija.
Dešerieva, T. I. (1999). Naxskie jazyki [Nakh languages]. In M. E. Alekseev (Ed.), Jazyki mira. Kavkazskie jazyki. [Languages of the World. Caucasian Languages.] (pp. 166–173). Moskva: Rossijskaja akademija nauk.
Dmitriev, N. K. (1940). Grammatika kumyksogo jazyka [Kumyk grammar]. Moskva: Akademija.
Forker, D. (2020). A grammar of Sanzhi Dargwa. Berlin: Language Science Press. Retrieved from https://langsci-press.org/catalog/book/250
Ibragimov, G. X. (2004). Rutulʹskij jazyk [Rutul]. Maxačkala: Narody Dagestana.
Kibrik, A. E. (1977). Opyt strukturnogo opisanija arčinskogo jazyka. Tom II. Taksonomičeskaja grammatika [Structural description of Archi. Volume II. Taxonomic grammar]. Moskva: Izdatelʹstvo Akademii nauk SSSR.
Kibrik, A. E. (1999). Élementy caxurskogo jazyka v tipologičeskom osveščenii [Elements of Tsakhur grammar in a typological perspective]. Nasledie: Moskva.
Kibrik, A. E., Kodzasov, S. V., & Olovjannikova, I. P. (1972). Fragmenty grammatiki xinalugskogo jazyka [Fragments of Khinalug grammar]. Moskva: Izdatelʹstvo Akademii nauk SSSR.
Magomedova, P. T. (2004). Bagvalinsko-russkij slovarʹ [Bagvalal-Russian dictionary]. Maxačkala: IJaLI.
Magometov, A. A. (1963). Kubačinskij jazyk [Kubachi]. Tbilisi: Akademija Nauk Gruzinskoj SSR.
Magometov, A. A. (1970). Agulʹskij jazyk [Aghul]. Tbilisi: Mecniereba.
Maxmudova, S. M. (2002). Grammatičeskie klassy slov i grammatičeskie kategorii rutulʹskogo jazyka [Grammatical word classes and grammatical categories of Rutul] (PhD thesis). Dagestan State University.
Molochieva, Z. (2010). Tense, aspect and mood in Chechen (PhD thesis). Universität Leipzig.
Musaev, K. M. (2018). Nogajsko-russkij slovarʹ [Nogai-Russian dictionary]. Moscow: Nauka.
Sumbatova, N. R., & Lander, J. A. (2014). Darginskij govor selenija Tanty: Grammatičeskij očerk. Voprosy sintaksisa. Moskva: Jazyki slavjanskoj kulʹtury.
Talibov, B. B. (2007). Buduxskij jazyk [Budukh]. Moskva: Akademija.
Žirkov, L. I. (1955). Lakskij jazyk [Lak]. Moskva: Akademija.

  1. Molochieva (2010: 254) defines Chechen ma- as a clitic, i.e. a syntactically independent but phonologically dependent morpheme.↩︎